LAUSR

OBJECTIVE To identify reliable predictors of periprocedural intensive care unit (ICU) admission after transvenous lead extraction (LE) in a high-volume center. DESIGN Retrospective observational study. SETTING University tertiary-care hospital. PARTICIPANTS All patients undergoing LE at San Raffaele Scientific Institute, Milan, Italy, from 2005 to 2015. INTERVENTIONS LE procedures were performed in the electrophysiology laboratories with a cardiac operating room on standby between the end of the morning surgical case and before the start of the afternoon surgical case. Most procedures were conducted with the patient under procedural sedation and analgesia. After LE, patients were admitted either to the ward or to the ICU. Medical history and intraprocedural data were recorded. MEASUREMENTS AND MAIN RESULTS Of the 389 procedures performed during the study period, 50 patients (13%) were admitted to the ICU owing to persistent hemodynamic instability or intraoperative complications requiring endotracheal intubation. Complete procedural success was achieved in 370 patients (95%), and the clinical success rate was 98.4%. No deaths were recorded. Five complications requiring emergency surgery (1.3%) were reported. Preprocedural right ventricular dysfunction (odds ratio (OR) 7.41; confidence interval 1.85-29.7; p < 0.01) and the need for general anesthesia (OR 12; confidence interval 1.49-97.06; p = 0.019) were independent predictors of ICU admission. CONCLUSIONS Preoperative identification of patients who need ICU admission after LE is crucial to increase patient safety and decrease hospital costs. Severe right ventricular dysfunction and need for general anesthesia identify patients with low cardiac reserve who are at increased risk for ICU admission after the procedure.